scaling-up post harvest management interventions in grain...

MARCH 2019

AGRA

Scaling-up Post Harvest Management Innovations

for grain legumes in Africa

– Study in Mozambique FINAL REPORT

MARCH 2019

AGRA

Scaling-up Post Harvest

Management Innovations for grain legumes in Africa

– Study in Mozambique FINAL REPORT

ADDRESS COWI Moçambique, Lda.

Ave. Zedequias

Manganhela, 95, 1st floor

P.O.Box 2242

Maputo

Mozambique

TEL +258 21 358 300

FAX +258 21 307 369

WWW cowi.co.mz

PROJECT NO. DOCUMENT NO.

16022-A 5

VERSION DATE OF ISSUE DESCRIPTION PREPARED CHECKED APPROVED

1 15.03.2019 Technical Report B. Cunguara,

IPCA, Ma N.

Dengo, M.

Tuominen,

Benedito

Cunguara

IPCA

SCALING-UP POST HARVEST MANAGEMENT INTERVENTIONS IN GRAIN LEGUMES - STUDY IN MOZAMBIQUE 5

https://d.docs.live.net/d9d63cbb82e262f4/Results and Evaluations/Market Access Program/Scaling up Post Harvest Management Innovation Technologies for Grain Legumes in

Mozambique.docx

CONTENTS

Executive Summary 7

1 Introduction 10

2 The Research problem 12

3 Study methodology 14

4 Study findings on the two key research

questions 16

4.1 What is the most effective way of accelerating the adoption of soybean post-harvest

management technologies? 16

4.2 What is the economic and social impact of increased use of postharvest management

technologies (PHT)? 17

5 Limitations and challenges faced 19

6 Conclusions and recommendations 20

6 SCALING-UP POST HARVEST MANAGEMENT INTERVENTIONS IN GRAIN LEGUMES - STUDY IN MOZAMBIQUE


Mozambique.docx

List of acronyms

AGRA Alliance for a Green Revolution for Africa

BCR Benefit-Cost Ratio

IDRC International Development Research Centre

NPV Net Present Value

PHT Post harvest technology

TNS Technoserve



Mozambique.docx

Executive Summary

Low agricultural productivity in Mozambique is a major constraint to the growth

of the Mozambican agriculture sector. While productivity levels are low even by

sub-Saharan countries standards, additional post-harvest losses that farmers

experience during threshing and limited/poor storage options serve to make the

predicament worse.

This has led the Alliance for a Green Revolution in Africa (AGRA) and the

International Development Research Centre to start a three-year (2016-2018)

applied research project “Scaling up Postharvest management innovations for

Grain Legumes in Africa” to bring effective, field-tested post-harvest innovations

for increasing productivity and reducing post-harvest loss of soya bean to

smallholder farmers in Mozambique.

The objective of this study was twofold. First, the study sought to identify the

most effective way of accelerating the adoption of soybean post-harvest

management technologies (PHT) in upper Zambézia, northern Mozambique, more

specifically in Gurué district. The selected PHT was the threshing machine. The

second objective was to estimate the economic and social impact of increased use

of PHTs in terms of production, income, nutrition, women’s situation and youth

employment.

In order to address these objectives, a two period panel data of about 320 soy

farming households was developed between 2016 (baseline study) and 2018

(endline study). Between the two survey studies, a monitoring study was

conducted in 2017, applying qualitative research methods such as key informant

interviews, focus group discussions, field observations, and weighing of soybean

to assess soybean losses during threshing.

In addition to these methods, a cost-benefit analysis of the thresher machine was

performed, along with the calculation of a food diversity index and a composite

index on women empowerment in agriculture.

Surveyed households were grouped using two typologies, one indicating whether

or not the farmer used any thresher, and another classifying the farming

households over time (persistent adopters if they used the thresher in both

periods; disadopters if they used the thresher in 2016 but not in 2018; new



Mozambique.docx

adopters if they only used the thresher in 2018 but not in 2016; and non-adopters

if they have never used the thresher).

On what concerns the first research question, i.e. what is the most effective

way of accelerating the adoption of soybean post-harvest management

technologies?, it is noted that farmers are very positive about the thresher

machine and eager to use it. However, the adoption of thresher machines for

soybean is associated with both investment profitability and the financial capacity

to buy the thresher machine, or to rent it from a local threshing service provider.

The cost-benefit analysis showed that, despite the high acquisition cost of a

threshing machine, the investment in the thresher is found highly profitable in a

12-year life cycle with the thresher operating at almost full capacity (1500 kg /

day) during 6 months per year. The financial capability to make the required

investment of buying the thresher machine can limit the opportunity to expand

adoption. The cost of thresher machine was reported too high in all three annual

studies1.

On what concerns the second research question, i.e. What is the economic and

social impact of increased use of postharvest management technologies

(PHT)?, descriptive results of the study indicate that soy farming households who

use the thresher machine attain higher productivity levels as they use improved

inputs more frequently than non-users. As a result, they also market a higher

proportion of their total production, become wealthier and have a more diversified

food diet; comparatively to soy farming households who do not use the thresher.

One of the potential impacts of the introduction of the thresher machine, is the

creation of employment opportunities for local labour, mostly related to the

operation and maintenance of the thresher machine. This extends to local agro-

dealers and suppliers of mechanic services, who are sought for spare parts of the

machine as well as repair services.

Another impact of the thresher machine is the expansion of soy farming plots and

the provision of threshing services to other services. Soy is perceived as one of

the most profitable crops and therefore there is a motivation of increasing soy

production as soon as the household can efficiently manage the harvest and post-

harvest (sales) process.

One additional impact of the thresher machine is the freed labour time for both

men and women: mechanized threshing works 6.5 times faster than manual

threshing. Female household members are spared from the responsibility of

cooking and catering for the manual threshing labour force. The freed time is

applied in other activities both leisure and income-generating.

Finally, soy milk and other derivatives of soy appear to be well accepted locally.

In the long run, this may imply that an increased use of soy could diversify the

local diet and boost the nutrition level of local population.

1 In 2018 the acquisition cost of one TECAP threshing machine (full price without

the AGRA grant) was $7,390 (endline study report, 2018).



Mozambique.docx

In terms of recommendations to accelerate the adoption of PHT, it is argued that

more affordable thresher machines should be available in the market in order to

facilitate investment in thresher machine business. One option would be to

facilitate imports and the private sector entry into this market, to make thresher

machines available at a relatively lower cost. Additionally, the development of

secondary markets (markets of used thresher machines) and machinery-hire

markets may also contribute to strengthen markets and expand supply of more

affordable machines.

On the demand side, service provision on fee-for-use models (used in site) should

be expanded, as well as increasing information on availability of thresher services

possibly through information technologies.



Mozambique.docx

1 Introduction

The Alliance for a Green Revolution in Africa (AGRA) is a non-profit organization

working with African governments, donors, civil society, the private sector and

farmers to improve the productivity and incomes of poor smallholder farmers in

Africa. AGRA has been investing in Mozambique since 2007 through projects for

improving seed systems by funding the establishment of seed companies,

distribution channels via agro-dealers, supporting to the fertilizer industry and

the development of agro-dealers.

Low agricultural productivity in Mozambique is a major constraint to the growth

of the Mozambican agriculture sector. While productivity levels are low even by

sub-Saharan countries standards, additional post-harvest losses that farmers

experience during threshing and limited/poor storage options serve to make the

predicament worse.

This has led AGRA and the International Development Research Centre (IDRC)

to start a three-year (2016-2018) applied research project “Scaling up

Postharvest management innovations for Grain Legumes in Africa” to bring

effective, field-tested post-harvest innovations for increasing productivity and

reducing post-harvest loss of soya bean to smallholder farmers in Mozambique.

The following objectives describe the planned actions to meet this goal:

› Objective 1: Scale up innovative post-harvest technologies (PHT), specifically

threshers and PICS bags, to achieve meaningful impacts in the lives of

farmers, women and youth;

› Objective 2: Assess the effectiveness of selected delivery models for the

chosen innovations; and

› Objective 3: Synthesize and disseminate evidence and lessons from the

scaling efforts to catalyse the field and inform policy change and investment.

In Mozambique, the key innovation brought by the project was the thresher

machine (Objective 1).

The project was implemented by Technoserve (TNS) in Gurué District (Zambézia

Province) with a specific focus on soya. Following TNS' model of 'small

commercial farmer' (SCF), the project identified emerging farmers and/ or

trusted existing local small agribusinesses who are trained, supported and

equipped by the programme to deliver the necessary inputs, extension,

mechanisation and other services to the smallholder farmers on a commercial

basis.

As part of the three-year project, 20 female SCF in Gurué were equipped with

new threshing machines on a co-financed basis. Additionally, 15 existing

mechanised SCFs in Gurué were supported to deliver mechanised threshing

services, and the production and marketing of nutritious soy-based food

products was promoted.



Mozambique.docx

COWI Mozambique was contracted by AGRA to carry out three annual studies for

the research project from 2016 to 2018, namely the baseline (2016), monitoring

(2017) and endline (2018) studies.

The present document is the Final Report, which follows the base, monitoring

and endline studies already submitted to AGRA. The Final Report aims to answer

the two main research questions with evidence from the three annual studies.

For that, the report is structured into six sections. After the Executive Summary

and this introductory chapter, the rest of the report unfolds according to the

following structure:

› Chapter two, Research Problem: presents the two main research questions,

objectives and indicators that guided the studies undertaken by COWI as

part of the research project;

› Chapter three, Study methodology: describes the methodology undertaken

for the collection and analysis of the annual studies data;

› Chapter four, Study findings on the two research questions: provides the

key findings of the annual studies to answer the research questions; and

› Chapter five, Limitations and challenges: provides the main challenges

faced by the annual studies, and the solutions applied to overcome them;

› Chapter six, Conclusions and recommendations: presents the main

conclusions of the study, as well as relevant recommendations for decision

making with regards to post-harvest management.



Mozambique.docx

2 The Research problem

The study in Mozambique aimed to answer two key learning questions2:

1 What is the most effective way of accelerating the adoption of soybean and

cowpea post-harvest management technologies?

2 What is the economic and social impact of increased use of postharvest

management technologies (PHT)?

As such, the objective of this study was twofold. On the one hand, the study

sought to identify the most effective way of accelerating the adoption of soybean

post-harvest management technologies (thresher machine) in Gurué. On the

other hand, the study sought to estimate the economic and social impact of

increased use of PHTs in terms of production, income, nutrition, women’s situation

and youth employment.

The thresher machines provided in the frame of the AGRA grant were acquired as

a matching grant with support from Technoserve (TNS). The grant was 50% of

the total value, farmers had to co-participate with 10% of the total amount, and

the remaining 40% pertained to bank loans. Some farmers received both a

thresher and a tractor.

The research project was also guided by pre-defined project indicators, namely:

Average number of months of adequate household food provision;

Household dietary diversity index;

1 Income generated from target crop by target farmers;

2 Percentage of crop losses during storage;

3 Volume of target crop sold by supported farmers (in Meticais);

4 Value of target crop sold by supported farmers;

5 Percentage of household using PHT technologies and

6 Number of household beneficiaries using PH technologies.

The research questions and project indicators guided the three annual studies

(2016-2018) conducted by COWI, namely the baseline (2016), monitoring (2017)

and endline (2018) studies.

In essence, COWI assessed the impact and the outcome of the thresher machine

project. This, combined with TNS data at input/output level, as well as specialized

2 Terms of Reference of the project and Inception Meeting held in Nairobi, Kenya

in July 2016.



Mozambique.docx

studies on post-harvest losses conducted by AGRA/TNS, will inform the impact

evaluation of the Post-Harvest Management Innovations project in Mozambique.



Mozambique.docx

3 Study methodology

This section presents the methodology followed by the three annual studies in

Mozambique.

Data collection methods

In order to address the study objectives, a two period panel data of about 320

soy farming households was developed between 2016 and 2018. The sample was

distributed between Gurué district (240 households), considered the treatment

group served by the TNS intervention, and neighbouring Malema district in

Nampula province (80 households) considered the control group3.

Between the two surveys (2016 and 2018), a monitoring study was conducted in

2017. This study applied qualitative methods such as key informant interviews

with mechanized threshing service providers, the District Government, TNS and

other implementing partners; focus group discussions with soy farmers; field

observations and weighing of soybean to assess soybean losses during threshing.

The surveyed households were grouped into two typologies, being the first a

binary indicator of whether the farmer used any thresher; and the second a more

dynamic definition that looks at farming households over time. As a result,

surveyed farmers were classified into four categories:

1 persistent adopters if they used the thresher in both periods;

2 disadopters if they used the thresher in 2016 but not in 2018;

3 new adopters if they only used the thresher in 2018 but not in 2016; and

4 non-adopters if they have never used the thresher.

Data analysis methods

For the qualitative a qualitative analysis matrix was prepared, while the

quantitative data was entered into a database. The data obtained from the

quantitative and qualitative methods were analysed by the research team, with

the support of AGRA in two preliminary data discussion rounds.

The impact of the intervention (adoption of mechanized threshing) was assessed

by comparing between the four categories of surveyed farmers, age and gender

of household head, and other characteristics.

Regression based models such as regression and matching, sub-classification and

regression, and the doubly-robust model were also considered to estimate the

impact of the intervention based on matched observations of beneficiary and the

counterfactual group.

3 Malema also practices soya production and has similar agro ecological

conditions to those of Gurué.



Mozambique.docx

All analyses were carried out in Stata 14.1 software.

Financial analysis

A financial analysis of mechanical threshers in soybeans and other grains was

conducted for both the monitoring (2017) and endline (2018) studies. The

financial analysis was based on the following theoretical framework:

Benefit-Cost Ratio (BCR): BCR is calculated by dividing the sum of present

benefit worth of stream (12 years) is divided by the sum of present cost. BCR >

1, signals a profitable investment at a given discount rate. Assumed discount rate

is 15%, corresponding to Banco de Moçambique (Central bank) rate of lending to

commercial banks in November 2018. The discount rate represents the

opportunity cost of capital.

Net Present Value (NPV): NPV is compounded by deducting the present worth

of cost stream from present worth of benefits. NPV > 0, signals investments

financially profitable at a given discount rate.

Mathematic terms:

Where:

n = 12 years

t= 1,2, 3, ……n(120

Bt = Benefit stream in each year

Ct = Cost stream in each year

i = Discount rate = 15%



Mozambique.docx

4 Study findings on the two key research

questions

This section reflects upon the two research questions, using the key findings of

the three annual studies (2016-2018).

4.1 What is the most effective way of accelerating the adoption of soybean post-harvest management technologies?

All three annual studies found that, in general, farmers appear very positive

about the thresher machine and are eager to use it. However, as argued in the

endline study report (2018) based on the results of a cost-benefit analysis, the

adoption of thresher machines for soybean is associated with both investment

profitability and the financial capacity to buy the thresher machine, or rent it

from a local threshing service provider. In those years when the market price of

soy is low, farmers cannot afford renting a PHT but instead they thresh

manually.

The availability of relatively large markets demanding for mechanical thresher

services, such as the one in Gurué, may contribute to accelerate the adoption of

soybean and cowpea post-harvest management technologies. Thus it is

important to expand these services to farming areas/ markets with high

population density and relatively high agriculture production. The investment in

the thresher is found highly profitable in a 12-year life cycle with the thresher

operating at almost full capacity (1500 kg / day) during 6 months per year i.e.

the full harvesting season (endline study report, 2018).

On the other hand, the financial capability to make the required investment of

buying the thresher machine can limit the opportunity to expand adoption. The

cost of thresher machine was reported too high in all three annual studies4. To

facilitate more investment in thresher machine business, more affordable

thresher machines should be available in the market. One option would be to

facilitate imports and the private sector entry into this market, to make thresher

machines available at a relatively lower cost.

In addition, development of secondary markets (markets of used thresher

machines) and machinery-hire markets may also contribute to strengthen

markets and expand supply of more affordable machines.

On the demand side, service provision on fee-for-use models (in use on the

project site) should be expanded, as well as, increasing information on

availability of thresher services. ITC should be explored to better match the

supply of thresher services with demand from farmers who need the services.

4 In 2018 the acquisition cost of one TECAP threshing machine (full price without

the AGRA grant) was $7,390 (endline study report, 2018).



Mozambique.docx

4.2 What is the economic and social impact of increased use of postharvest management technologies (PHT)?

Descriptive results from the two surveys5 undertaken indicate that soy farming

households who use the thresher machine usually belong to households whose

head is older, and they usually have a non-farm income source. Participation in

non-farm income generating activities is usually positively and significantly

correlated with education, which implies that better educated households have

better access to PHTs. PHT users have better housing, although it is difficult to

establish causality.

In terms of production, PHT users attain higher productivity levels as they use

improved inputs more frequently than non-users. As a result, they also market a

higher proportion of their total production, and have a more diversified food diet,

both women and men.

Additionally, the study points that continuing or persistent adopters became

wealthier. Persistent adopters were able to expand their cropped area and animal

stocks, used improved inputs more frequently and therefore obtained higher

productivity levels. They also increased their incomes and nutritional status,

proxied by food diversification index. Non-adopters and disadopters show no

improvement in asset ownership.

The endline study (2018) found that it takes between 2-5 people to operate a

thresher machine, typically one operador who is the main responsible for the

machine itself, and 1-4 assistants who help feeding the machine with soy,

checking the bagging of clean grains and controlling that the remaining parts of

the plant are discarded.

Thus one of the potential impacts of the introduction of the thresher machine, is

the creation of employment opportunities for local labour, mostly related to the

operation and maintenance of the thresher machine. This extends to local agro-

dealers and suppliers of mechanic services, who are sought for spare parts of the

machine as well as repair services.

Another impact of the thresher machine identified in the three annual studies is

the expansion of soy farming plots and the provision of threshing services to other

services. According to the endline study report (2018), farmers that used PHTs

were able to expand their average cropped area from about 12 hectares in 2016

to about 21 hectares in 2018; compared to about 5 hectares among non-adopters

in 2018. The increase in landholding size was boosted by the production of

soybean seeds among PHT users. A recently constructed seed factory operates in

Magige, about 20 km to Gurué city. Many farmers debushed new fields to cultivate

soybeans, both grains and seeds.

Soy is perceived as one of the most profitable crops and therefore there is a

motivation of increasing soy production as soon as the household can efficiently

manage the harvest and post-harvest (sales) process.

5 Baseline study (2016) and Endline study (2019).



Mozambique.docx

One additional important impact of the thresher machine found in the three annual

studies, is that it frees labour time. Both men and women are involved in manual

threshing (baseline study report, 2017), and the use of PHT has a direct impact

on their time saved. According to the endline study (20186), mechanized threshing

works 6.5 times faster than manual threshing. Additionally, for female household

members in particular, the use of the thresher machine frees them from the

responsibility of cooking and catering for the labour force hired to perform manual

threshing. The freed time of both men and female is applied in other activities

both leisure and income-generating.

Finally, soy milk and other derivatives of soy appear to be well accepted locally.

In the long run, it is possible to assume that an increased use of soy could diversify

the local diet and boost the nutrition level of local population.

6 Based on TNS data from the Progress and Results of” Research Project on

Post-Harvest Losses in the Soybean Value Chain in Mozambique", 24/10/2018.



Mozambique.docx

5 Limitations and challenges faced

Several challenges were faced in each of the three annual studies.

The Baseline study was initially affected by the political instability that ran the

country in 2016, resulting in local farmers' mistrust towards the survey team.

Another challenge faced then was the mapping of soy farming areas in Malema,

in the absence of a mapped registry by the district authorities. Last but not

least, the availability of COPAZA member threshing service providers was in

some cases compromised, and those were interviewed by phone. In order to

mitigate these challenges, several identification measures were set up.

The main constraint faced in the Monitoring study was the soy weighting exercise

(conducted with harvested soybean in piles, rather than in bags). Additionally, it

was rather difficult to mobilize female thresher owners for the study, as they had

just recently adhered to the project and received the threshing machine.

Additional communication efforts were made with the husbands.

As for the observation of threshing in Malema, all observations made were for

manual threshing. It was not possible to observe mechanized threshing as there

are considerably fewer PHT users in Malema than in Gurué, due to the fact that

most the threshing machines used in Malema come from Gurué. During the field

visit mechanized threshing was still occurring in Gurué and was expected to

move to Malema within a few weeks' time, after data collection had ended. As

such, mechanized threshing was only observed in Gurué.

The main constraint faced in the Endline study was the high attrition rate due to

household mobility, linked to high commodity price volatility, especially for

soybeans, pigeon peas, and maize. In order to overcome this, replacements

were made with soy farmers from the same geographic unit, to reach the study

sample.



Mozambique.docx

6 Conclusions and recommendations

In order to accelerate the adoption of PHT, particularly mechanized threshing,

the overall study results suggest that the AGRA grant via TNS is very helpful for

the farmer, who would not be able to bear the full cost of the thresher (or the

combo of thresher and tractor). The mechanisms to accelerate the adoption of

PHTs should include a grant to farmers.

There is also a need to find ways to operate the machines for longer periods per

year, to increase its current profitability level. The threshing of other grains

(such as maize) and beans (such as cowpea) apart from soy, could play an

important role in this.

Additionally, it is recommended that more affordable thresher machines should

be available in the market in order to facilitate investment in thresher machine

business. One option would be to facilitate imports and the private sector entry

into this market, to make thresher machines available at a relatively lower cost.

Additionally, the development of secondary markets (markets of used thresher

machines) and machinery-hire markets may also contribute to strengthen

markets and expand supply of more affordable machines.

On the demand side, service provision on fee-for-use models (used in site) should

be expanded, as well as increasing information on availability of thresher services

possibly through information technologies.

scaling-up post harvest management interventions in grain...

Documents